Providing Novel Approach for Classification of Soil and Crop Prediction Using Decision Support System

Decision Support System has been applied to solve variety of agricultural problems. It provides the framework that allows both the decision-makers and farmers to make good decisions. Decision support systems (DSS) use databases, human– machine to combine a large number of models to realize scientific decision-making. The environmental data plays a vital role in agriculture decision, which changes at a rapid rate. Keeping these data updated can be done by using a Service Oriented Approach (SOA). The goal of the irrigation system is to determine the exact amount of needed water and the exact timing for applying it also the soil type required for particular crop. The amount of water applied is determined dependent on each user situation. The system can use to build a decision-making model to for irrigation for different users. Decision tree is a well known approach for classification in data mining. C4.5 and Classification and Regression Trees (CART) are two widely used decision tree algorithms for classification. The main drawback of C4.5 algorithm is that, it is biased towards attributes with more values while CART algorithm produces misclassification errors when the domain of the target attribute is very large. In view of these limitations, this paper presents a modified decision tree algorithm. The test proves that the modified decision tree algorithm has higher classification accuracy when compared to C4.5 and CART algorithms.